FORM-REGULATION IN CERIANTHUS. llj 



only the entoderm of the oesophageal region but the great mass 

 of mesenterial differentiations aboral to the oesophagus, /. c., the 

 chief digestive region of the entoderm. There is little doubt 

 that processes of secretion are much more active and persistent 

 in this region than in the oesophageal entoderm, and consequently 

 the quantity of fluid reaching the enteron either in connection 

 with the secretory process or by osmosis through the body-wall 

 in consequence of the presence of soluble substances secreted 

 into the enteron is greater than in cesophageal pieces and disten- 

 sion continues for a longer time. Whatever the exact explana- 

 tion may be the fact is patent in all cases. As will appear below,, 

 it is perhaps of considerable importance in connection with the 

 appearance of heteromorphosis. 



The pieces of this series were prepared September 26, 1902, 

 All pieces were observed at intervals after the operation and 

 among the methods of closure were found those represented 

 diagrammatically in longitudinal section in Figs. 19 and 26. The 

 body-wall and the oesophagus are indicated as in previous fig- 

 ures. In Fig. 26 the aboral end of the body-wall has bent over 

 orally much farther than in Fig. 19. In both cases, however, 

 the margin of the body-wall has united on all sides with that of 

 the oesophagus, thus forming a closed piece composed partly of 

 oesophagus and partly of body-wall. The enteric cavity of this 

 piece is of course the enteric cavity of the lateral half of the body 

 in the region from which it was taken. In Figs. 19 and 26 the 

 external surface of the oesophageal portion on the right of the 

 figure is ectodermal, but in the normal animal formed the axial 

 surface of the oesophagus. The only essential difference between 

 such pieces and cesophageal pieces (Child, 'o^d] consists in the 

 greater length of the body-wall in the former. 



Tentacle-regeneration at the oral ends of the pieces occurred 

 in the typical manner (Figs. 20 and 27), each producing about 

 half the number of tentacles possessed by the parent specimen, a 

 result to be expected from the number of intermesenterial cham- 

 bers contained in each piece. 



About two months later (November 20) two of the pieces 

 showed tentacles beginning to develop on the aboral ends. One 

 of these pieces is shown in Fig. 17. The tentacles on the oral 



